By primer extension, we determined the complete nucleotide sequence of the gene encoding the outer capsid protein VP4 of human rotavirus strain K8. The VP4 gene is 2,359 base pairs in length with 5'- and 3'-noncoding regions of 9 and 25 nucleotides, respectively. The gene contains a long open reading frame of 2,325 bases capable of coding for a protein of 775 amino acids. When compared to other human rotaviruses, the VP4 of the strain K8 had an insertion of one amino acid after the l35th residue as found in simian rotavirus strains, and, in addition, it had a deletion of one amino acid (575th residue). The VP4 gene of K8 was only 60-70% homologous in amino acid sequence to the VP4 of other virulent human rotaviruses. This is an unusual observation since the VP4s of all other virulent human rotavirus strains exhibit greater than 90% amino acid homology. In contrast, the VP7 amino acid sequence of the K8 strain was quite similar to that of other serotype 1 human rotaviruses (greater than 98% homology). In our previous study in which the cross-reactive neutralization epitopes on VP4 of human rotaviruses were analyzed, K8 was found to bear unique VP4 neutralization epitopes. This strain, which belongs to subgroup II and serotype 1, was not neutralized by any of six anti-VP4 neutralizing monoclonal antibodies which reacted with serotype 1, 3 and 4, or serotype 1 through 4 human rotavirus strains. Thus, the K8 strain appears to have a unique VP4 gene previously not described.